What happens to your receipt.

Lunch Party is built and operated by Spergs, LLC, a Florida limited liability company. You can reach us at privacy@spergs.com for anything privacy-related, or to point out something on this page that isn’t accurate.

We don’t sell, rent, or trade your data, and we don’t run ads. There is no advertising business here to feed.

No account. There’s no sign-up, no login, no email, no password. We don’t know who you are and we don’t try to find out.

Your receipt photo is turned into text and discarded. The image is uploaded to our server, run through our own text-recognition service (it does not go to a third-party OCR cloud), turned into the line items you see, and not written to any database or receipt history.

Nothing about your split is stored server-side. The order you build (items, the people you add, their tip preferences, their payment handles) lives in your browser’s localStorage on your device. Each request to our server carries the order with it and the server keeps nothing between requests; there is no database. The in-progress split survives closing the tab so you can pick up where you left off. Clear it from the app (starting a new scan does it), wipe your browser’s site data, or use a private window (which discards everything when the window closes).

We don’t process payments. The “pay back” buttons are ordinary deep links to Venmo or Zelle. Money moves through them, not through us.

The receipt image

• When you scan or upload a receipt, the image is sent to our server and passed to our self-hosted text-recognition pipeline (Tesseract behind a job queue, all on our own infrastructure).
• It is processed to extract the text, then dropped. We don’t keep a copy, we don’t build a receipt archive, and there is no “your past receipts” feature because there is nowhere it could come from.
• Transient processing can mean the image briefly exists on disk as a temporary file while a worker reads it; those are not retained as part of any record and aren’t tied to anything identifying you.

The order you build

• The parsed line items and prices, any edits you make, the people you add (the names you type; make them whatever you like), each person’s tip preference, who was assigned which items, and the computed totals.
• Optionally, a Venmo or Zelle handle per person, used to build the prefilled payment link. It’s only useful if you type it in; you don’t have to.
• All of this is held in your browser’s localStorage on your device (under the key lunch-party-workflow-state). It is sent to our server on each step only so the server can parse, total, and return the result, and it is not persisted there. The in-progress split survives closing the tab, so you can resume where you left off; starting a new scan in the app clears it, or you can wipe your browser’s site data, or use a private window (everything is discarded when the private window closes).

Connection logs

• We don’t run application-level analytics. The reverse proxy in front of the app (nginx) keeps standard access logs that can include connection IP addresses and timestamps, the way most web servers do. Those are operational, rotated, and not joined to any account, because there are no accounts.

• No accounts, so no email address, password, or profile.
• No analytics, behavioural tracking, session replay, or third-party advertising pixels. There is no tracking SDK on the page.
• No contact-book or address-book access. The people in a split are names you type.
• No payment data. We never see card numbers, bank details, or whether a Venmo/Zelle payment actually happened.
• No receipt history. There is no database; we couldn’t show you a past split if we wanted to.

The text-recognition step is not a third party. It’s our own service on our own infrastructure, which is the whole reason the receipt doesn’t leave our control. The third parties that exist are:

• Venmo / Zelle. When you tap a “pay back” link, your device opens Venmo or Zelle with the amount prefilled. From that point you’re in their app under their privacy policy; we’re not part of the transaction and don’t hear back about it.
• Hosting. The site and app run on a server we operate. Standard hosting operations mean the underlying-infrastructure provider has physical access to the hardware; they don’t have credentialed access to the service.

We don’t share data with anyone else, because there’s very little to share and nowhere it’s kept.

In your browser: the in-progress split (items, people, assignments, totals) for as long as the tab is open.

On our server: only while a request is being handled. The receipt-parsing and calculation pipeline is stateless. Each step receives the full order, returns a result, and retains nothing. There is no database storing splits, receipts, or people.

• Your split is held in your browser’s localStorage on your device, not on our servers. To delete it: start a new scan in the app (which clears it), wipe your browser’s site data for lunchparty.net / app.lunchparty.net, or use a private window (everything is discarded when the window closes).
• The receipt image is dropped after text extraction; any transient temp file is short-lived and not kept as a record.
• Reverse-proxy access logs follow normal operational rotation and aren’t tied to a split or a person.

The strongest privacy control here is structural: there’s no account to manage, no server-side history to clear, and no profile to export, because none of that is created in the first place. Your in-progress split lives in your own browser, under your control. If you’d rather not put a real name or a payment handle in, don’t. The split still works with “Person 1, Person 2.”

If you have a question about this policy, or you’re in a jurisdiction (UK/EU GDPR, California CCPA) and want to ask what we hold on you, email privacy@spergs.com. The answer is almost always “nothing on our servers; your split lives in your own browser storage until you clear it,” but we’ll engage with any request in good faith and try to respond within 30 days.

Lunch Party isn’t directed at children and isn’t intended for use by anyone under 13. We don’t knowingly collect anything from children, and the design collects very little from anyone. If you believe a child has sent us something through this service, email privacy@spergs.com.

If something material changes (particularly if we ever add an account system, persistent storage, or a third-party service), we’ll update the “Last updated” date above and describe what changed. We won’t quietly start collecting more than this page says.

Spergs, LLC
privacy@spergs.com